Back to results
US 11,629,961 B2Granted
Crosswalk detection
AssigneeGM Cruise Holdings LLC
InventorsHarman Kumar, Juan Fasola, Yu-Cheng Lin
Life of this patent
Estimating the term…
Abstract
Systems and methods are disclosed for detecting crosswalk locations. Crosswalk locations can be detected by determining at least two painted lines on a road surface, and then combining the at least two painted lines on the road surface into a grouping of related elements. The grouping of related elements can be classified into a crosswalk (or in some embodiments a type of crosswalk) based on attributes of the grouping of related elements, where the attributes of the grouping of the elements includes a distance between the at least two painted lines, and an orientation of the grouping of related elements on the road surface.
Claims20
A computer-implemented method for crosswalk detection comprising:
- determining, from sensor data acquired from one or more sensors of an autonomous vehicle (AV), one or more clusters of the sensor data representative of a plurality of features of an environment along a navigated route;
- processing the one or more clusters of the sensor data to identify at least one feature of the plurality of features in the environment, wherein the at least one feature is identified based on a spatial closeness of the one or more clusters;
- generating by grouping the one or more clusters of the sensor data to form at least two painted lines on a road surface identified from the plurality of features;
- combining the at least two painted lines on the road surface into a grouping of related elements associated with the one or more clusters of the sensor data, the at least two painted lines comprising a minor axis and a major axis;
- identifying the grouping of related elements as a candidate zebra-type crosswalk or a parallel line-type crosswalk when the spatial closeness of the one or more clusters meets a predefined threshold;
- classifying the candidate zebra-type crosswalk or the parallel line-type crosswalk as a zebra-type crosswalk or the parallel line-type crosswalk based on attributes of the grouping of related elements, the attributes of the grouping of the related elements including a distance between the at least two painted lines, and an orientation of the grouping of related elements on the road surface with respect to an angle of travel of the autonomous vehicle;
- identifying an inconsistency of the classifying the zebra-type crosswalk or the parallel line-type crosswalk from a set of pre-mapped data associated with the navigated route;
- updating the pre-mapped data based on the classifying the zebra-type crosswalk or the parallel line-type crosswalk;
- and
- controlling the direction of the movement of the autonomous vehicle during navigation, based on the updating of the pre-mapped data.
7 more elements
A non-transitory computer readable medium comprising instructions stored thereon, the instructions when executed cause at least one processor to:
- determine, from sensor data acquired from one or more sensors of an autonomous vehicle (AV), one or more clusters of the sensor data representative of a plurality of features of an environment along a navigated route,
- process the one or more clusters of sensor data to identify at least one feature of the plurality of features in the environment, wherein the at least one feature is identified based on a spatial closeness of the one or more clusters;
- generate by grouping the one or more clusters of sensor data to form at least two painted lines on a road surface identified from the plurality of features;
- combine the at least two painted lines on the road surface into a grouping of related elements associated with the one or more clusters of sensor data, the at least two painted lines comprising a minor axis and a major axis;
- identify the grouping of related elements as a candidate zebra-type crosswalk or a parallel line-type crosswalk when the spatial closeness of the one or more clusters meets a predefined threshold;
- classify the candidate zebra-type crosswalk or the parallel line-type crosswalk as a zebra-type crosswalk or the parallel line-type crosswalk based on attributes of the grouping of related elements, the attributes of the grouping of the related elements including a distance between the at least two painted lines, and an orientation of the grouping of related elements on the road surface with respect to an angle of travel of the autonomous vehicle, with respect to the angle of the travel of the autonomous vehicle;
- identify an inconsistency of the classifying the zebra-type crosswalk or the parallel line-type crosswalk from a set of pre-mapped data associated with the navigated route;
- update the pre-mapped data based on the classifying the zebra-type crosswalk or the parallel line-type crosswalk;
- and
- control the direction of the movement of the autonomous vehicle during navigation, based on the updating of the pre-mapped data.
7 more elements
A computing system comprising:
- at least one non-transitory computer readable medium comprising instructions stored thereon, wherein the instructions when executed cause the computing system to:
- determine, from sensor data acquired from one or more sensors of an autonomous vehicle (AV), one or more clusters of the sensor data representative of a plurality of features of an environment along a navigated route;
- process the one more clusters of sensor data to identify at least one feature of the plurality of features in the environment, wherein the at least one feature is identified based on a spatial closeness of the one or more clusters;
- generate by grouping the one or more clusters of sensor data to form at least two painted lines on a road surface identified from the plurality of features;
- combine the at least two painted lines on the road surface into a grouping of related elements associated with the one or more clusters of sensor data, the at least two painted lines comprising a minor axis and a major axis;
- identify the grouping of related elements as a candidate zebra-type crosswalk or a parallel line-type crosswalk when the spatial closeness of the one or more clusters meets a predefined threshold;
- classify the candidate zebra-type crosswalk or the parallel line-type crosswalk as a zebra-type crosswalk or the parallel line-type crosswalk based on attributes of the grouping of related elements, the attributes of the grouping of the related elements including a distance between the at least two painted lines, and an orientation of the grouping of related elements on the road surface with respect to an angle of travel of the autonomous vehicle;
- identify an inconsistency of the classifying the zebra-type crosswalk or the parallel line-type crosswalk from a set of pre-mapped data associated with the navigated route;
- update the pre-mapped data based on the classifying the zebra-type crosswalk or the parallel line-type crosswalk;
- and
- control the direction of the movement of the autonomous vehicle during navigation, based on the updating of the pre-mapped data.
8 more elements
Dependencies are inferred from the claim text.